1756-EN2TR DLR Ring Faults

Hi Everyone , Need some advice.

I have a ControlLogix 1756-L71 with 2 EN2TR Communication Cards, both of which are configured as Ring supervisors for two different Device Level rings.

While one network/ring is operating perfectly fine with no faults.

The other network card throws up an error frequently on the network tab: 'Rapid Fault/Restore Cycles'. The port configuration data doesn't seem pretty. I have attached the screenshots for reference.

What I observed after resetting these counters was that the FCS Errors and MAC Transmit/Receive Errors were increasing, the port config data few hours after resetting are as such:

Port 1:
FCS Errors: 17, MAC Transmit Error: 1776, MAC Receive Error: 10

Port 2:
FCS Errors: 32908, MAC Transmit Error: 0, MAC Receive Error: 25483

I have about 40 nodes on this network.

I've checked the relevant manual and the possible reasons for the errors:
1)Intentional disconnections
2)Duplex Mismatch between devices
3)Electromagnetic Noise
4)Wrong cables/connections.

I've checked, the only plausible cause seems to be noise. Even that, I've used Cat6 twisted Ethernet cable which is fairly(according to me ) isolated.

My question is that can Electromagnetic noise create this many errors or is it something more obvious?? And what do FCS and MAC errors mean and imply ?

Thanks.

What does your topology look like? Specifically, you're not trying to put a switch into the middle of the ring are you? That doesn't work well and you'll need to install ETAPs if it is the case.

Next, duplex settings are absolutely imperative to monitor on these ring setups make sure you are matched on every device. In my own experience I have had better luck forcing everything to 100/full but your mileage may vary.

Last, with that amount of fcs errors, my first guess would be a media issue if you don't have a switch in there like I mentioned before. Are you using shielded cable? If so, make sure only one end is bonded so you're not inadvertently causing ground loops. Good luck.

thumperbs said:

Last, with that amount of fcs errors, my first guess would be a media issue if you don't have a switch in there like I mentioned before. Are you using shielded cable? If so, make sure only one end is bonded so you're not inadvertently causing ground loops. Good luck.

Click to expand...

Are you sure that cat.6 or cat.5e specifies one end connection on the shield? I don't believe so.

Anyway, if the cabling is certified, you could expect that everything with it is ok.

thumperbs said:

What does your topology look like? Specifically, you're not trying to put a switch into the middle of the ring are you? That doesn't work well and you'll need to install ETAPs if it is the case.

Next, duplex settings are absolutely imperative to monitor on these ring setups make sure you are matched on every device. In my own experience I have had better luck forcing everything to 100/full but your mileage may vary.

Last, with that amount of fcs errors, my first guess would be a media issue if you don't have a switch in there like I mentioned before. Are you using shielded cable? If so, make sure only one end is bonded so you're not inadvertently causing ground loops. Good luck.

Click to expand...

Thanks,
Its a basic ring topology with a 1756-EN2TR as supervisor connected to about 40 odd nodes(same device). Pretty straight forward ring without any switch or any other networking device.

I've checked almost twice the configuration of all the nodes on the ring, all are auto-negotiated and operating at 100 Mbps, Full duplex.

I've used the an Unshielded Cat6 cable, but I hardly think it is susceptible to electrical noise. Besides, its the same cable I've used for the other network and it seems fine. Even then I am still considering it, but how do I know this for sure ??
Using a shielded cable maybe an obvious answer
But I'd have to change a heck lot of cables.

Is there any SIMPLE way to check if there is any noise on the network ??

'Rapid Fault/Restore Cycles' indicates that you had 5 or more DLR faults in 30 seconds.

Media counters indicate corrupted data and I suspect that you have one of the cables bad.
I can't recommend shielded cables, DLR should work just fine with unshielded cables.
Instead I would start with just 2 or 3 nodes and watch media counters. they should not increment. Then add one node at the time and monitor media counters.
Is there a chance that you have at least one node that is not DLR capable? Or switch in the middle? Or a media converter? These all can bring your DLR down.

EN2TR rapid ring fault

I have faced the rapid ring fault and checked the devices connected to the network and changed some device ethernet cables. BUt problem solved by changing EN2TR module . Port 2 got failure which is not communicating even with single device .

If you have less devices with RPI rate more then 250 ms you can manage with port 1 on same EN2TR.